Data-driven and model-guided systematic framework for media development in CHO cell culture

Exploring metabolic effects of dipeptide feed media on CHO cell cultures by in silico model-guided flux analysis

There is a growing interest in perfusion or continuous processes to achieve higher productivity of biopharmaceuticals in mammalian cell culture, specifically Chinese hamster ovary (CHO) cells, towards advanced biomanufacturing. These intensified bioprocesses highly require concentrated feed media in order to counteract their dilution effects. However, designing such condensed media formulation poses several challenges, particularly regarding the stability and solubility of specific amino acids. To address the difficulty and complexity in relevant media development, the biopharmaceutical industry has recently suggested forming dipeptides by combining one from problematic amino acids with selected pairs to compensate for limitations. In this study, we combined one of the lead amino acids, L-tyrosine, which is known for its poor solubility in water due to its aromatic ring and hydroxyl group, with glycine as the partner, thus forming glycyl-L-tyrosine (GY) dipeptide. Subsequently, we investigated the utilization of GY dipeptide during fed-batch cultures of IgG-producing CHO cells, by changing its concentrations (0.125 × , 0.25 × , 0.5 × , 1.0 × , and 2.0 ×). Multivariate statistical analysis of culture profiles was then conducted to identify and correlate the most significant nutrients with the production, followed by in silico model-guided analysis to systematically evaluate their effects on the culture performance, and elucidate metabolic states and cellular behaviors. As such, it allowed us to explain how the cells can more efficiently utilize GY dipeptide with respect to the balance of cofactor regeneration and energy distribution for the required biomass and protein synthesis. For example, our analysis results uncovered specific amino acids (Asn and Gln) and the 0.5 × GY dipeptide in the feed medium synergistically alleviated the metabolic bottleneck, resulting in enhanced IgG titer and productivity. In the validation experiments, we tested and observed that lower levels of Asn and Gln led to decreased secretion of toxic metabolites, enhanced longevity, and elevated specific cell growth and titer. KEY POINTS: • Explored the optimal Tyr dipeptide for the enhanced CHO cell culture performance • Systematically analyzed effects of dipeptide media by model-guided approach • Uncovered synergistic metabolic utilization of amino acids with dipeptide.

Driving towards digital biomanufacturing by CHO genome-scale models

Genome-scale metabolic models (GEMs) of Chinese hamster ovary (CHO) cells are valuable for gaining mechanistic understanding of mammalian cell metabolism and cultures. We provide a comprehensive overview of past and present developments of CHO-GEMs and in silico methods within the flux balance analysis (FBA) framework, focusing on their practical utility in rational cell line development and bioprocess improvements. There are many opportunities for further augmenting the model coverage and establishing integrative models that account for different cellular processes and data for future applications. With supportive collaborative efforts by the research community, we envisage that CHO-GEMs will be crucial for the increasingly digitized and dynamically controlled bioprocessing pipelines, especially because they can be successfully deployed in conjunction with artificial intelligence (AI) and systems engineering algorithms.

Inducible transgene expression in CHO cells using an artificial transcriptional activator with estrogen-binding domain

Biopharmaceuticals, including therapeutic antibodies, are rapidly growing products in the pharmaceutical market. Mammalian cells, such as Chinese hamster ovary (CHO) cells, are widely used as production hosts because recombinant antibodies require complex three-dimensional structures modified with sugar chains. Recombinant protein production using mammalian cells is generally performed with cell growth. In this study, we developed a technology that controls cell growth and recombinant protein production to induce recombinant protein production with predetermined timing. Expression of green fluorescent protein (GFP) gene and a single-chain antibody fused with the Fc-region of the human IgG1 (scFv-Fc) gene can be induced and mediated by the estrogen receptor-based artificial transcription factor Gal4-ERT2-VP16 and corresponding inducer drugs. We generated CHO cells using an artificial gene expression system. The addition of various concentrations of inducer drugs to the culture medium allowed control of proliferation and transgene expression of the engineered CHO cells. Use of 4-hydroxytamoxifen, an antagonist of estrogen, as an inducing agent yielded high gene expression at a concentration more than 10-fold lower than that of β-estradiol. When scFv-Fc was produced under inducing conditions, continuous production was possible for more than 2 weeks while maintaining high specific productivity (57 pg cell-1 day-1 ). This artificial gene expression control system that utilizes the estrogen response of estrogen receptors can be an effective method for inducible production of biopharmaceuticals.

Debottlenecking and reformulating feed media for improved CHO cell growth and titer by data-driven and model-guided analyses

Designing and selecting cell culture media along with their feeding are a key strategy to maximize culture performance in biopharmaceutical processes. However, the sensitivity of mammalian cells to their culture environment necessitates specific nutritional requirements for their growth and the production of high-quality proteins such as antibodies, depending on the cell lines and operational conditions employed. In this regard, previously we developed a data-driven and in-silico model-guided systematic framework to investigate the effect of growth media on Chinese hamster ovary (CHO) cell culture performance, allowing us to design and reformulate basal media. To expand our exploration for media development research, we evaluated two chemically defined feed media, A and B, using a monoclonal antibody-producing CHO-K1 cell line in ambr15 bioreactor runs. We observed a significant impact of the feed media on various aspects of cell culture, including growth, longevity, viability, productivity, and the production of toxic metabolites. Specifically, the concentrated feed A was inadequate in sustaining prolonged cell culture and achieving high titers when compared to feed B. Within our framework, we systematically investigated the major metabolic bottlenecks in the tricarboxylic acid cycle and relevant amino acid transferase reactions. This analysis identified target components that play a crucial role in alleviating bottlenecks and designing highly productive cell cultures, specifically the addition of glutamate to feed A and asparagine to feed B. Based on our findings, we reformulated the feeds by adjusting the amounts of the targeted amino acids and successfully validated the effectiveness of the strategy in promoting cell growth, life span, and/or titer.

Cell-culture process optimization via model-based predictions of metabolism and protein glycosylation

In order to meet the rising demand for biologics and become competitive on the developing biosimilar market, there is a need for process intensification of biomanufacturing processes. Process development of biologics has historically relied on extensive experimentation to develop and optimize biopharmaceutical manufacturing. Experimentation to optimize media formulations, feeding schedules, bioreactor operations and bioreactor scale up is expensive, labor intensive and time consuming. Mathematical modeling frameworks have the potential to enable process intensification while reducing the experimental burden. This review focuses on mathematical modeling of cellular metabolism and N-linked glycosylation as applied to upstream manufacturing of biologics. We review developments in the field of modeling cellular metabolism of mammalian cells using kinetic and stoichiometric modeling frameworks along with their applications to simulate, optimize and improve mechanistic understanding of the process. Interest in modeling N-linked glycosylation has led to the creation of various types of parametric and non-parametric models. Most published studies on mammalian cell metabolism have performed experiments in shake flasks where the pH and dissolved oxygen cannot be controlled. Efforts to understand and model the effect of bioreactor-specific parameters such as pH, dissolved oxygen, temperature, and bioreactor heterogeneity are critically reviewed. Most modeling efforts have focused on the Chinese Hamster Ovary (CHO) cells, which are most commonly used to produce monoclonal antibodies (mAbs). However, these modeling approaches can be generalized and applied to any mammalian cell-based manufacturing platform. Current and potential future applications of these models for Vero cell-based vaccine manufacturing, CAR-T cell therapies, and viral vector manufacturing are also discussed. We offer specific recommendations for improving the applicability of these models to industrially relevant processes.

Data-driven prediction models for forecasting multistep ahead profiles of mammalian cell culture toward bioprocess digital twins

Recently, the advancement in process analytical technology and artificial intelligence (AI) has enabled the generation of enormous culture data sets from biomanufacturing processes that produce various recombinant therapeutic proteins (RTPs), such as monoclonal antibodies (mAbs). Thus, now it is very important to exploit them for the enhanced reliability, efficiency, and consistency of the RTP-producing culture processes and for the reduced incipient or abrupt faults. It is achievable by AI-based data-driven models (DDMs), which allow us to correlate biological and process conditions and cell culture states. In this work, we provide practical guidelines for choosing the best combination of model elements to design and implement successful DDMs for given hypothetical in-line data sets during mAb-producing Chinese hamster ovary cell culture, as such enabling us to forecast dynamic behaviors of culture performance such as viable cell density, mAb titer as well as glucose, lactate and ammonia concentrations. To do so, we created DDMs that balance computational load with model accuracy and reliability by identifying the best combination of multistep ahead forecasting strategies, input features, and AI algorithms, which is potentially applicable to implementation of interactive DDM within bioprocess digital twins. We believe this systematic study can help bioprocess engineers start developing predictive DDMs with their own data sets and learn how their cell cultures behave in near future, thereby rendering proactive decision possible.

Enhanced cell growth, production, and mAb quality produced in Chinese hamster ovary-K1 cells by supplementing polyamine in the media

Polyamines such as putrescine (PUT), spermidine (SPD), and spermine (SPM) are amine group-containing biomolecules that regulate multiple intracellular functions such as proliferation, differentiation, and stress response in mammalian cells. Although these biomolecules can be generated intracellularly, lack of polyamine-synthesizing activity has occasionally been reported in a few mammalian cell lines such as Chinese hamster ovary (CHO)-K1; thus, polyamine supplementation in serum-free media is required to support cell growth and production. In the present study, the effects of biogenic polyamines PUT, SPD, and SPM in media on cell growth, production, metabolism, and antibody quality were explored in cultures of antibody-producing CHO-K1 cells. Polyamine withdrawal from media significantly suppressed cell growth and production. On the other hand, enhanced culture performance was achieved in polyamine-containing media conditions in a dose-dependent manner regardless of polyamine type. In addition, in polyamine-deprived medium, distinguishing metabolic features, such as enriched glycolysis and suppressed amino acid consumption, were observed and accompanied by higher heterogeneity of antibody quality compared with the optimal concentration of polyamines. Furthermore, an excessive concentration of polyamines negatively affected culture performance as well as antibody quality. Hence, the results suggest that polyamine-related metabolism needs to be further investigated and polyamines in cell growth media should be optimized as a controllable parameter in CHO cell culture bioprocessing. KEY POINTS: • Polyamine supplementation enhanced cell growth and production in a dose-dependent manner • Polyamine type and concentration in the media affected mAb quality • Optimizing polyamines in the media is suggested in CHO cell bioprocessing.

Genome-scale modeling of Chinese hamster ovary cells by hybrid semi-parametric flux balance analysis

Flux balance analysis (FBA) is currently the standard method to compute metabolic fluxes in genome-scale networks. Several FBA extensions employing diverse objective functions and/or constraints have been published. Here we propose a hybrid semi-parametric FBA extension that combines mechanistic-level constraints (parametric) with empirical constraints (non-parametric) in the same linear program. A CHO dataset with 27 measured exchange fluxes obtained from 21 reactor experiments served to evaluate the method. The mechanistic constraints were deduced from a reduced CHO-K1 genome-scale network with 686 metabolites, 788 reactions and 210 degrees of freedom. The non-parametric constraints were obtained by principal component analysis of the flux dataset. The two types of constraints were integrated in the same linear program showing comparable computational cost to standard FBA. The hybrid FBA is shown to significantly improve the specific growth rate prediction under different constraints scenarios. A metabolically efficient cell growth feed targeting minimal byproducts accumulation was designed by hybrid FBA. It is concluded that integrating parametric and nonparametric constraints in the same linear program may be an efficient approach to reduce the solution space and to improve the predictive power of FBA methods when critical mechanistic information is missing.